Measuring and modelling of landfill gas emissions

Abstract

Landfills start emitting landfill gas (LFG) shortly after waste placement and can continue for more than twenty years after the placement of the final cover and landfill closure. The concentration of LFG components will vary significantly over this period; therefore, in order to evaluate long-term impacts on the landfill workers, the surrounding public and the environment, a means to measure and model these emissions were developed in this research.

Flux chambers have been utilized for several years to measure the flux of various gases from soil surfaces. Many of the flux chambers have had problems with the pressure differential between the chamber and the atmosphere that has resulted in underestimation or overestimation of the true flux. The flux chamber developed in this research minimizes the pressure differential with a pressure transduced, process controller and peristaltic pump i a feedback loop. Laboratory evaluation of the system found that the pressure differentials were minimized with an average range of -0.036 to 0.009 mm H2) over five experiments with nearly complete gas recovery (99%).

Flux measurements with the flux chamber performed at 3 different landfills exhibited excellent pressure control. The flux measurements of CH4, TCE, PCE and H2S obtained at the various landfills were dependent on the cover condition. The flux of CH4 was 1 to 3 orders of magnitude lower from an intact cover than the equivalent flux measured at holes in the cover.

The Landfill Emission Simulation (LES) model, developed in this research simulated the production, transport and emission of CH4, CO2, vinyl chloride and H2S from the landfill surface. The model evaluated the emission during the landfill construction until a user defined length after landfill closure. Contributions of holes and CH4 oxidation in cover on the flux of these gases were also simulated. Oxidation in the top 0.3 m of the final cover was found to reduce the methane flux and increase the carbon dioxide flux.

Concentrations of vinyl chloride and H2S at off-site points (100 m, 200 m, 500 m, 1000 m) were predicted using the Gaussian Plume model and the emissions of vinyl chloride and H2S generated from the LES model. These concentrations for the site investigated did not exceed the MOE 1/2 hour point of impingement standard.